The Efficacy and Safety of Knotless Barbed Sutures in the Surgical Field: A Systematic Review and Meta-analysis of Randomized Controlled Trials

The knotless barbed suture is an innovative type of suture that can accelerate the placement of sutures and eliminate knot tying. However, the outcomes of previous studies are still confounding. This study reviewed the application of different types of barbed sutures in different surgeries. We searched PubMed, EMBASE, CENTRAL and ClinicalTrials.gov to identify randomized controlled trials (RCTs) addressing the application of barbed sutures up to Feb. 2015. Two reviewers independently screened the literature and assessed the risk of bias of included studies. Then meta-analysis was performed using RevMan 5.3 software. Sensitivity analysis and subgroup analysis was performed. Seventeen RCTs (low to moderate risk of bias) involving 1992 patients were included. Compared with conventional sutures, the barbed suture could reduce suture time (SMD=−0.95, 95%CI −1.43 to −0.46, P = 0.0001) and the operative time (SMD=−0.28, 95%CI −0.46 to −0.10, P = 0.003), not significantly increase the estimated blood loss (SMD=−0.09, 95%CI −0.52 to 0.35, P = 0.70), but could lead to more postoperative complications (OR = 1.43, 95%CI 1.05 to 1.96, P = 0.03), These results varied in subgroups. Thus, barbed sutures are effective in reducing the suture and operative time, but the safety evidences are still not sufficient. It need be evaluated based on special surgeries and suture types before put into clinical practice.

Scientific RepoRts | 6:23425 | DOI: 10.1038/srep23425 we present the available evidence in terms of the efficacy and safety of different types of knotless barbed sutures in different surgeries by performing a systematic review and meta-analysis of the current literature.
Of the 17 trials, 16 trials were performed using computer-generated randomization, 1 used the coin toss; 9 performed allocation concealment through central randomization; 5 applied blinding only to patients and 1 was open labeled; and 4 applied blinding to outcome assessors while 1 did not. The loss to follow-up occurred in 0 to 14.1% of patients. In general, the risk of bias was low to moderate in RCTs (Supplementary Table 1).
Publication bias. Publication bias was assessed using Begg's funnel plots. The shape of the funnel plots appeared symmetric in the barbed vs. conventional suture, suggesting no evidence of publication bias ( Supplementary Figures 10-13).

Discussion
Generally, barbed sutures reduced the suture time in nearly all types of surgeries, as well as the operative time.
Although barbed sutures resulted in more postoperative complications, no significant change occurred concerning the estimated blood loss. Moreover, the results differed in different surgeries, and the bidirectional barbed suture appeared to be better than the unidirectional barbed suture. To eliminate interference from confounding factors, we performed subgroup analysis by surgeries and barbed type, and the results were varied. First, our subgroup results showed a significant association between suture time and barbed suture in 5 types of surgeries (laparoscopic myomectomies, cosmetic surgeries, sacrocolpopexies, gastric bypasses and robot-assisted laparoscopic prostatectomies). Taken together, these findings suggested that the barbed suture significantly shortened the suture time in laparoscopic myomectomies (5.50 min), cosmetic surgeries (6.76 min), sacrocolpopexies (13.60 min), gastric bypasses (11.30 min) and robot-assisted laparoscopic prostatectomies (0.10 min). Thus the effectiveness need be evaluated based on particular surgeries.
In addition, although the overall effect of operative time decreased in barbed groups, a subgroup analysis suggested that only the operative time of laparoscopic myomectomies (2.73 min) and gastric bypasses (11.70 min) were significantly reduced, which was partially consistent with previous studies 3, 21 Furthermore, a subgroup analysis also indicated that the use of barbed sutures resulted in less blood loss in laparoscopic myomectomies, which differed from results obtained in a previous study 21 .
Regarding the postoperative complications, the subgroup analysis only indicated that the number of cosmetic surgeries was higher in the barbed suture groups than the control, whereas the pooled results obtained from other surgeries or studies reported no difference. This result may be due to the two studies 14,17 of cosmetic surgeries, both of which had dermal closure performed on one side with the barbed suture and the conventional suture on the opposite side, which increased the risk of surgical site infection. Moreover, previous studies concerning gynecological surgeries reported that bowel obstruction might be attributable to the increased risk of either adhesions or inflammation caused by the barbs entrapped in the novel suture 3,21 .
Another concern our meta-analysis focused on is the comparison of different barbed suture types. Compared with the conventional suture, a unidirectional barbed suture decreased the suture and operative times significantly and also demonstrated more postoperative complications, whereas the pooled results of a bidirectional barbed suture did not statistically differ from the control in all outcomes. Thus, the bidirectional barbed suture appeared safer than the unidirectional sutures; although the pooled overall effect indicated no difference. Interestingly, the sensitivity analysis also showed no differences in postoperative complications between the control and either of the barbed groups. The most probable explanation for this result may be that the unidirectional barbed suture required more skillful surgeons. Because such sutures require cuts and re-stitches once suturing errors occurred, this can probably cause more damage to human tissue. Nevertheless, regarding the bidirectional barbed suture, when the barbs in one direction are in the wrong locations, then it can be modified using the other direction to maintain the tension.
Although there are three types of barbed suture commercially available, this study only identified research studies concerning the unidirectional barbed and bidirectional barbed suture; there were no RCTs on humans referring to the third type, Stratafix (STRATAFIX Knotless Tissue Control Devices, Ethicon Inc., Somerville, NJ, USA). Thus, the feasibility and safety among different barbed sutures used in in vivo studies should be taken into consideration in the future 22 .
In addition to the favorable outcomes described above from pooled results, numerous other benefits of barbed sutures exist regardless of the patients or surgeons. For example, the barbed suture can eliminate knot tying and the speed of the placement of the sutures. Furthermore, eliminating the need for an assistant's hand to follow the  suture placement, enhancing the equal distribution of tension, and creating the possibility of improved scar cosmoses are also compelling validations for using this state-of-the-art technique.
Our pooled outcome provides convincing evidence for the relationship between the barbed suture and some important surgical indicators. However, caution should be taken to explain the pooled results due to the limitations of our study. (1) Relatively high heterogeneity among studies was estimated for surgical related outcomes, particularly in suture time and estimated blood loss. (2) Although our literature search was extensive, it did not    cover conference publications and letters to the editor. (3) There was a lack of cost-effectiveness, cost-benefit, and cost-utility analyses, and the descriptive economic analysis of this study was imperfect. (4) Considering the high heterogeneity of all of the research studies, we performed the SMD for most of the outcomes.
Nevertheless, our results renew a latest meta-analysis on barbed sutures. To the best of our knowledge, this is the most comprehensive meta-analysis to date investigating the association between barbed and traditional sutures.
In conclusion, with the advantages of shorter suture and operative times, postoperative complications were likely to occur more often when using unidirectional barbed sutures. Future studies should also be performed to comprehensively analyze the effect on cost-effectiveness.

Methods
Study identification and selection. The MEDLINE, EMBASE and the Cochrane Library databases were searched using the following terms: "barbed" OR "knotless" AND "suturing" OR "suture" (last updated in Feb. 2015). To modify the results and to avoid publication bias, we also searched clinical trials registered in ClinicalTrials.gov (last updated in Feb. 2015).
All studies had to meet the following inclusion criteria: (a) study design had to be a RCT based on human subjects; (b) patients underwent surgical operation; (c) interventions had to be conventional suture vs. barbed suture; and (d) studies should report at least one of the outcomes with detailed data, such as suture time, estimated blood loss, operative time, and postoperative complications. The following exclusion criteria were also applied: (a) conventional sutures were other materials, such as mesh or staple rather than smooth sutures; (b) abstracts or overlapped studies; and (c) studies published in languages other than English. The computer search was supplemented with manual searches for references of included studies.

Data Extraction and Outcome Measures.
We imported the search results into bibliographic citation management software (EndNote X7, Thomson Reuters, USA). Two reviewers independently collected the data and reached a consensus on all items. The following items were extracted from each study if available: first author's surname, publication year, original country, sample size, type of suture, and postoperative complications.
The main outcome measures chosen for the current meta-analysis were operative time, suture time, estimated blood loss or change in hemoglobin level and postoperative complications. Heterogeneity of the outcomes was assessed to confirm the appropriateness of combining individual studies.

Definition.
Operative time was defined as the total time of surgery. Suture time was defined as the time needed for the completion of the surgical site incision, anastomosis time, and closure time. Estimated blood loss (ml) or change in hemoglobin level (g/dL) (different studies reported different indices of blood loss) was defined as the blood loss during the operation, and it was usually obtained from both the anesthesia records and/or the surgeons' operative reports. After surgeries, postoperative complications of the suture were also recorded. Both unidirectional and bidirectional barbed sutures were evaluated together as the barbed suture category.

Methodological Quality Assessment. The risk of bias of included RCTs and was assessed following
Cochrane recommendations, considering random sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessment, incomplete outcome data and selective reporting 23 . We searched the protocol of each trial to assess the selective reporting. Publication bias was evaluated using the funnel plot.
Data Synthesis and Analysis. The studies were divided into seven subgroups according to the seven different surgeries, which were also divided into two subgroups according to the two types of barbed suture; in addition, separate meta-analysis was performed within different subgroups. In all analyses, we estimated the pooled mean difference (MD) and standardized mean difference (SMD) to assess continuous data, while the pooled odds ratios (ORs) were calculated for the assessment of dichotomous data (postoperative complications). The pooled estimations regarding outcomes expressed as either dichotomous or continuous variables were calculated using the random effect model (postoperative complications using fixed effect model). The existence of statistical heterogeneity between the included studies was assessed using the χ 2 test and I 2 test. In addition, we also performed sensitivity analyses to examine the robustness of the estimates and assessed the risk of publication bias using Begg's funnel plots. For all analyses, P < 0.05 was considered statistically significant. Statistical analyses were performed using the software programs Review Manager (Version 5.3).